Hi,
I have a two cluster system, with a drbd and a postgresql using this drbd.
My prefered node is 'mylogintest1'.
When mylogintest goes standy, the failover works wonderful, but when I
switch it on again,
the heartbeat tries to demote the drbd before the postresql (and the
filesystem) is stop,
so it fails and it can not go to master on mylogintest1.
Here my colocation and order rules.
What is wrong? Do I need another rule for postgres? Or must I group
postgres and filesystem?
Or is the sequence in the cib-file relevant?
<constraints>
<rsc_colocation id="colDRBDIP" rsc="msDRBD" rsc-role="Master"
score="INFINITY" with-rsc="resIP"/>
<rsc_colocation id="colFSDRBD" rsc="resFilesystem" score="INFINITY"
with-rsc="msDRBD" with-rsc-role="Master"/>
<rsc_order first="resIP" id="ordIPDRBD" then="msDRBD"
then-action="promote"/>
<rsc_order first="msDRBD" first-action="promote" id="orderDRBDFS"
then="resFilesystem" then-action="start"/>
<rsc_order first="resFilesystem" id="ordFilesysPostgres"
then="resPostgres"/>
<rsc_order first="resIP" id="ordIPApache" then="resApache"/>
<rsc_order first="resApache" id="ordApacheTomcat" then="resTomcat"/>
<rsc_colocation id="colPostgresFS" rsc="resPostgres" score="INFINITY"
with-rsc="resFilesystem"/>
<rsc_colocation id="colApacheIP" rsc="resApache" score="INFINITY"
with-rsc="resIP"/>
<rsc_colocation id="colTomcatApache" rsc="resTomcat" score="INFINITY"
with-rsc="resApache"/>
<rsc_location id="locIP" node="mylogintest1" rsc="resIP"
score="INFINITY"/>
</constraints>
Greetings
Jochen
--
Dr. rer. nat. Jochen Lienhard
UB Universität Freiburg
Dezernat EDV
Tel: (0761) 203-3908
Email: [email protected]
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
